Welcome back to the Launch Key 🚀
Last week I crossed an idea off my potential list of businesses I wanted to start. Saying ‘No” is important. But it’s still a good idea for someone else. This week: how it could work, what to charge, and the numbers if you want to go after it.
Let's get into it.
Gmail users may wish to read online since some parts may be clipped.
Comment below or answer this weeks poll and help improve this newsletter.
Table of Contents
Pull To Eject
Last week I told you about an idea (Reverse Bucket List) I keep rejecting: an AI back office for the home-service trades. HVAC, plumbing, electrical, pool service, landscaping, pest control, cleaning.
In the Spring, I wrote that a few roofers were moving first, using AI to find hail damage and send warm leads straight to their sales teams. I recently read how Tyler Griffin uses Grok bot to run his HVAC and plumbing businesses (SwiftPro Heating, Cooling & Plumbing). Other outliers are experimenting, but most of the trades haven't caught up. That gap is the opportunity.
The plumber is under a sink
Call five local service companies at 4:30 on a Tuesday afternoon. Count how many pick up.
Invoca, a call-tracking company, found that 27% of calls to home-services businesses go unanswered. That isn't laziness. The owner is on a roof. The tech is under a sink. The office manager is out sick.
Most of those callers don't leave a voicemail. They call the next company in their search results.
Speed matters even when someone does call back. A 2011 Harvard Business Review study tested 2,241 companies with web leads. Only 37% responded within an hour, and 23% never responded at all. The ones that answered within the hour were nearly seven times more likely to qualify the lead than companies that waited even one more hour.
Alex Hormozi tells a story about a business owner who converts 55% of his leads. His secret, as the story goes: he pays his aunt $60,000 a year to do one thing. When a lead comes in, she drops whatever she's doing and calls it.
That job can now be done by software, around the clock, for a fraction of the cost.
They know. They're stuck.
Here's the opening. Only about one in five US businesses uses AI at all, according to the Census Bureau. Among the smallest firms, adoption isn't growing.
The trades are no different. ServiceTitan surveyed more than 1,000 residential contractors this year and found only 25% use AI. The top reasons? It's too complicated to set up (53%), and they don't trust what it produces (43%).
Read that again. They aren't saying they don't need it. They're saying they don't have time to figure it out and don't trust it to talk to their customers.
That's not a technology problem. It's a service you can sell.
What you actually sell
Don't sell "AI." Nobody who owns a pest control company wants AI. They want fewer missed jobs and fewer evenings doing paperwork.
So sell three things for one monthly fee:
Every call answered. An AI receptionist picks up 24/7, answers the basic questions, and books the appointment.
Every quote followed up. Automatic texts and emails go out after a quote, so the "let me think about it" customers hear back without anyone remembering to send them something.
A calendar that stays full. Reminders, rescheduling, and a review request after every job.
Web and marketing can come later. Start with the phone. That's where the money leaks out.
The numbers
On Oct 6 I showed you how AI changes the economics of a service business: Payroll costs less than $250 a month. The short version: the biggest change isn't the margin percentage, it's breakeven. Back in May I put it another way: AI is the new junior layer. With no payroll, it takes far fewer clients to build something substantial.
Here's how that looks for this business. These are illustrative numbers, not a quote.
What owners pay now: an in-house receptionist runs about $38,000 a year in wages alone (BLS, May 2025), before benefits. Off-the-shelf AI answering tools run roughly $200 to $2,000+ a month, and the owner still has to set them up and manage them.
What you charge: a flat monthly fee, say $1,000 to $1,500, for the whole back office, set up and managed. Greg Isenberg's rule is to price against the person you're replacing, not against your own costs.
What you need: 10 clients at $1,000 a month is $120,000 a year. Your tool costs per client are a fraction of that.
Compare that with my old web firm. We needed to generate $100K+ in revenue per employee. Here there are zero employee costs to cover.
One rule from the Oct 6 issue matters even more: charge a flat fee, not hourly. If you bill by the hour, the time AI saves goes to the client. With a flat fee, you keep it.
Why this belongs to someone with grey hair
The tech is the easy part. Anyone can sign up for an AI receptionist this afternoon.
The hard part is sitting across from a 58-year-old HVAC owner who's been burned by three marketing agencies and getting him to trust you with his phone line. That takes thirty years of business meetings. It takes knowing how a small business really runs. It takes being the person he already knows from Rotary, the golf course, or church.
Then you get better with every client. Isenberg puts it well: "A competitor can download the model in an afternoon. They can't download three hundred jobs' worth of your mistakes."
I live in the fastest-growing county in North Carolina. Every new subdivision needs lawns cut, pests controlled, and AC fixed in August. There’s rarely a plumbing call that’s not an emergency. The trades can't hire office staff fast enough. That's the opening.
Where it goes wrong:
Selling the tool instead of the result. "Booked jobs" sells. "Voice AI agent" doesn't.
Trying to serve every trade. Pick one trade in one town. Plumbers talk to plumbers. Your second client comes from your first.
Building everything custom. Build one setup and repeat it. Every new client should take less work than the one before. I can’t tell you how many times I messed this one up in the past.
No human backup. The AI will eventually say something wrong to a customer. Decide who gets the phone call when that happens, and tell the client before it does.
What you can do this week
Do the 4:30 test. Call five local service businesses near the end of the workday and write down how many answer, how many send you to voicemail, and how many call back within an hour.
If most of them miss the call, you've just done your market research. And if one of them is a business you already know, offer to fix it for free for a month. Your first client is your proof.
Bet on yourself. One trade, one town, one client at a time.
Now go launch something 🚀
Work on your business, not in it.
Become An AI Expert In Just 5 Minutes
If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.
This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.
Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.
Old School Wisdom
The E-Myth Revisited — Michael E. Gerber (1995, original 1986)
Gerber's argument is that most small businesses are started by technicians, people who are good at the work, who then end up with a job they can't quit instead of a business. His fix is to build the business like a franchise prototype: a system that produces the same result every time, whoever runs it.
In 1995, building that system meant operations manuals, org charts, and staff to train. Gerber said you needed a franchise prototype. He just didn't know it could run on software.
That's the model for this week's business. Your first client is your prototype. Write down every step. By client five, you shouldn't be inventing anything.
Modern Tools
Spend 15 minutes reading Greg Eisenberg’s $100B AI-native services opportunity article.
Free Knowledge
Founders #293: Ray Kroc — David Senra
Kroc was a 52-year-old milkshake-machine salesman when he saw what the McDonald brothers had built and turned their system into a franchise. Gerber uses McDonald's as the original franchise prototype. Senra tells the story better than anyone. It's a good drive's worth of listening for anyone thinking about a repeatable service business.
Recommendations
🚏 The Cautionary : Learn to challenge conventional wisdom.
📕 MGMT Playbook : Practical management insights straight to your inbox every Wednesday.
🏢 Retirepreneur : Your guide to turning decades of experience into profitable consulting, coaching, or course businesses, without risking your retirement security.
🗃 Dealroom Business Success Uncovered : Learn directly from billionaire entrepreneurs on how to grow a business. Join a community of 2,000+ innovators.
🧾 Redefining Retirement : Too Young to Retire, Too Old To Take Orders? Yeah, Me Too.
Visual Crapshoot

What should year 4 of The Launch Key do more of?
Launch Key readers – thank you for your support and feedback. I appreciate each and every one of you as I work to build something you value.
Remember, if there's anything you'd like to share — a recommendation, a story idea, or just a note to say hi, hit the reply button and fire away.






